Advanced Intelligent Computing Technology and Applications : 20th International Conference, ICIC 2024, Tianjin, China, August 5–8, 2024, Proceedings, Part IV / edited by De-Shuang Huang, Zhanjun Si, Chuanlei Zhang.

Format:

Book

Contributor:

Huang, De-Shuang, editor.

Si, Zhanjun, editor.

Zhang, Chuanlei, editor.

Series:

Lecture Notes in Artificial Intelligence, 2945-9141 ; 14878

Language:

English

Subjects (All):

Artificial intelligence.

Computers.

Computer networks.

Data mining.

Image processing--Digital techniques.

Image processing.

Computer vision.

Software engineering.

Artificial Intelligence.

Computing Milieux.

Computer Communication Networks.

Data Mining and Knowledge Discovery.

Computer Imaging, Vision, Pattern Recognition and Graphics.

Software Engineering.

Local Subjects:

Artificial Intelligence.

Computing Milieux.

Computer Communication Networks.

Data Mining and Knowledge Discovery.

Computer Imaging, Vision, Pattern Recognition and Graphics.

Software Engineering.

Physical Description:

1 online resource (514 pages)

Edition:

1st ed. 2024.

Place of Publication:

Singapore : Springer Nature Singapore : Imprint: Springer, 2024.

Summary:

This 6-volume set LNAI 14875-14880 constitutes - in conjunction with the 13-volume set LNCS 14862-14874 and the 2-volume set LNBI 14881-14882 - the refereed proceedings of the 20th International Conference on Intelligent Computing, ICIC 2024, held in Tianjin, China, during August 5-8, 2024. The total of 863 regular papers were carefully reviewed and selected from 2189 submissions. The intelligent computing annual conference primarily aims to promote research, development and application of advanced intelligent computing techniques by providing a vibrant and effective forum across a variety of disciplines. This conference has a further aim of increasing the awareness of industry of advanced intelligent computing techniques and the economic benefits that can be gained by implementing them. The intelligent computing technology includes a range of techniques such as Artificial Intelligence, Pattern Recognition, Evolutionary Computing, Informatics Theories and Applications, Computational Neuroscience & Bioscience, Soft Computing, Human Computer Interface Issues, etc.

Contents:

Intro

Preface

Organization

Contents - Part IV

Intelligent Fault Diagnosis

Desirable Properties Based Neural Network Explanations Evaluation Method for Fault Diagnosis

1 Introduction

2 Related Work

2.1 Explanation Method

2.2 Explanation Evaluation Method

3 Method

3.1 Dummy Feature

3.2 Class Sensitive

3.3 Repeated Feature

3.4 Feature Coalition

4 Results and Discussion

4.1 CWRU Dataset

4.2 PU Dataset

4.3 Discussion

5 Conclusion and Future Work

References

Variational Autoencoder and Graph Attention Root Cause Localization Model Based on Log Data and Graph Structure

2 System Log Processing and Log Dependency Graph

2.1 System Log Processing

2.2 Log Dependency Graph

3 Model Framework

3.1 Construction of LDG

3.2 Feature Extraction

3.3 Feature Aggregation and Classification

4 Experiment and Result Analysis

4.1 Datasets and Environment

4.2 Evaluation Metrics

4.3 Analysis of Results

5 Conclusion and Prospects

An Interpretable Fault Prediction Method Based on Machine Learning and Knowledge Graphs

2 Literature Review

2.1 Predictive Maintenance

2.2 Explainable Predictive Maintenance

3 Methodology

3.1 Framework

3.2 Fault Prediction Model

3.3 Explainable Methods

3.4 Turbofan Engine Knowledge Graph

4 Fault Prediction and Interpretation

4.1 Fault Prediction Modeling Experiments

4.2 Experiment

5 Conclusions

STRCA: A Lightweight and Accurate Root Cause Analysis System Based on 5G Signalling Trace

2 STRCA Design

2.1 5G Signalling Parsing

2.2 Trace Anomaly Detection

2.3 Anomaly Network Element Root Cause

3 Evaluation

3.1 Performance of Trace Anomaly Detection

3.2 Performance of Root Cause Localization.

3.3 Time Complexity Analysis

4 Conclusion

Fault Diagnosis of Rotating Equipment Unbalance Problem Based on Denoising Stacked Autoencoders

2 Methodology

2.1 Preliminary

2.2 Model Architecture

3 Experiments

3.1 Dataset Description

3.2 Implementation Details

4 Results

4.1 Supervised Learning in Five Classification

4.2 Supervised Learning in Binary Classification

4.3 Ablation Studies

5 Discussion

6 Conclusion

ConvNeXt-BiGRU Rolling Bearing Fault Detection Based on Attention Mechanism

2.1 Wavelet Packet Decomposition to Extract Signal Features

2.2 ConvNeXt-BiGRU Network Messaging Process

2.3 Structure of the Fault Detection Model

3 Experiment

3.1 Choice of Datasets

3.2 Parameters of the Fault Detection Model Structure

3.3 Analysis and Verification Through Experimentation

3.4 Model Assessment and Training Index

PBAFS: Preference-Based Active Feature Selection for Fault Diagnosis and Prevention of HVAC Systems

2 Methods and Technical Routes

2.1 Problem Formulation

2.2 A Preference-Based Feature Selection Algorithm

2.3 A Preference-Based Feature Selection Algorithm

3 Data Experiment

3.1 Experimental Data of a Practical Chiller System

MicroDACP: Microservice Fault Diagnosis Method Based on Dual Attention Contrastive Learning and Graph Attention Networks

3.1 Anomaly Detection Method Based on Dual Attention Mechanism

3.2 Fault Root Cause Localization Method Based on Graph Attention Networks

4 Experiments

4.1 Evaluation of Indicators

4.2 Baseline Methods

4.3 Evaluation Results

5 Conclusion

References.

SpikeFusionNet: A Hybrid Approach to Robotic Fault Diagnosis Using Spiking Neural Dynamics

2.1 CNNs in Fault Diagnosis

2.2 SNNs in Fault Diagnosis

3 Methods

3.1 The Structure of SpikeFusionNet

3.2 HistAware-LIF Neuron

3.3 Neural Temporal-Intensity Coding

3.4 Similarity Learning

4 Simulation Experiments

4.1 Data Description and Preprocessing

4.2 Evaluation Indicators

4.3 Experimental Results and Analysis

4.4 Ablation Experiment

Non-stridden Convolution and Bidirectional Cross-Scale Features Fusion Network for Steel Surface Defect Detection

2 Methodologies

2.1 Review of RT-DETR

2.2 Proposed Network

3 Experimental Results and Analysis

3.1 Dataset and Experimental Environment

3.2 Evaluation Metrics

3.3 Analysis of Experimental Results

4 Conclusions

A Lightweight Physics-Informed Neural Network Model Based on Causal Discovery for Remaining Useful Life Prediction

3 Experiments and Results

3.1 Experimental Setup

3.2 Evaluation Criteria

3.3 Setting of the Proposed Network Structure

3.4 Experimental Results and Performance Evaluation

An Effective Deep SVM Approach for Fault Diagnosis of 25 Hz Track Circuit

1.1 Normal Operating State

1.2 Fault Operating State

2 CNN-LSTM-Attention-SVM Fusion Model

2.1 Notations

2.2 Model Layer

3 Simulation Experiment

3.1 Experimental Data

3.2 Experimental Results and Analysis

Interpretable Remaining Useful Life Prediction Based on Causal Feature Selection and Deep Learning

3.2 Evaluation Criteria.

3.3 Setting of the Proposed Network Structure

3.4 Empirical Findings and Performance Analysis

Natural Language Processing and Computational Linguistics

Large Models and Multimodal: A Survey of Cutting-Edge Approaches to Knowledge Graph Completion

2 KGC Based on Representation Learning

2.1 Methods Based on Translation

2.2 Methods Based on Tensor Decomposition

2.3 Methods Based on Neural Network

2.4 Methods Based on Graph Neural Network

3 KGC Based on a Large Model

3.1 Methods Based on Pre-training Models

3.2 Methods Based on LLMs

4 KGC Based on Multimodal Learning

5 Conclusion and Prospect

5.1 Enhance Semantic Consistency of Multimodal Fusion

5.2 Enhancing the Capture of Long-Term Dependencies Between Nodes

5.3 Utilizing LLMs for Smarter KGC

Legal-LM: Knowledge Graph Enhanced Large Language Models for Law Consulting

2 Approach

2.1 Overview

2.2 Keyword Extraction for Legal Issues

2.3 Legal-LM Pre-training

2.4 Legal-LM Fine-Tuning

2.5 Legal-LM Direct Preference Optimization

2.6 Retrieval Augmentation

3.1 Data Setup

3.2 Training Setup

3.3 Evaluation Protocols

3.4 Result

3.5 Analysis

Knowledge Graph Reasoning for Few-Shot Problems

2 Relate Work

3 Work

3.1 GCN Pre-training Embeddings

3.2 Reinforcement Learning Reasoning

3.3 Meta-learning Training

4 Experiment

4.1 Datasets

4.2 Comparative Models

4.3 Evaluation Metrics

4.4 Experimental Results and Analysis

Label-Related Adaptive Graph Construction Based on Attention for Multi-label Text Classification

3 Preliminaries

4 Label-Related Adaptive Graph Construction Method.

4.1 Label-Related Text Representation

4.2 Graph Construction

4.3 Gated Fusion Strategy

4.4 Multi-label Text Classification

5 Experiment

5.1 Experiment Setting

5.2 Experimental Results and Analysis

5.3 Ablation Test

6 Conclusion and Future Work

TCGA: A Grid-Tagging NER Model Enhanced by Fusing Position and Region Information

3.1 Encoder Module

3.2 Convolution Module

3.3 Co-predictor Module

3.4 Decoding

3.5 Learning

4 Experiment Setting

4.1 Implementation Details

4.2 Experimental Results

4.3 Model Ablation Studies

COVID-19 Rumor Detection Based on Heterogeneous Graph Convolutional Network with Cross-Domain Contrastive Learning

2.1 Rumor Detection Methods

2.2 Contrastive Learning

3.1 Problem Statement

3.2 Overall Architecture

3.3 Heterogeneous Graph Construction Module

3.4 Rumor Detection Module

3.5 Cross-Domain Contrastive Learning Module

4.2 Experimental Settings

4.3 Baselines

4.4 Results and Analysis

4.5 Ablation Analysis

4.6 Case Study

A Novel World Knowledge Aware Universal Representations of Design Patterns Based on Pre-trained Language Models

2 Method

3 Experiment Setup

4 Results and Discussions

4.1 RQ1: How Does Knowledge in Pre-trained Natural Language Models Affect Model Performance?

4.2 RQ2: Does Adaptation Training in the Design Pattern Corpus Contribute to the Language Model Ability to Model the Association Between Natural Language and Design Pattern?.

4.3 RQ3: Is the Language Model-Based Approach Better than the Word-Based Approach in the Task of Modeling the Association Between Natural Language and Design Patterns?.

ISBN:

981-9756-72-3